COVID-19 and Thromboembolism: FAQs

What is the proposed pathophysiology of venous thromboembolism as a complication of COVID-19?

COVID-19 triggers all three arms of Virchow’s triad: endothelial injury, hypercoagulability, and blood flow stasis.³

• COVID-19 may increase levels of von Willebrand factor and Factor VIII via endothelial injury.³
• Release of inflammatory cytokines (cytokine storm) could activate the coagulation cascade.² Antiphospholipid antibodies may play a role.² On autopsy, megakaryocytes have been found in unusually high numbers outside the bone marrow (e.g., in the lungs and heart).⁴
• Immobility, and treatments used for seriously ill COVID-19 patients such as fluid restriction and high PEEP, may cause blood flow stasis and microthrombi.³

COVID-19-induced hypoxia facilitates thrombus formation.¹

Some drugs being used as treatments for COVID-19 may increase thrombosis risk directly (e.g., baricitinib), or indirectly by reducing efficacy of antithrombotics (e.g., tocilizumab could potentially speed metabolism of oral anticoagulants).^2,13,14

Severely ill COVID-19 patients may have non-COVID-19-specific contributors to VTE risk, such as central lines.²

DIC has been reported, but it is unclear if this is related to a specific effect of COVID-19, or a nonspecific complication of critical illness.² Contrary to what is usually seen in DIC, COVID-19 coagulopathy is characterized by normal or even increased fibrinogen.¹⁰ Moreover, overt bleeding seems not to be common in COVID-19 patients.¹⁰

How does COVID-19-associated thromboembolism present clinically?

In a German cohort of 12 autopsied patients (52 to 87 years of age) who died with a confirmed case of COVID-19, microthrombi were common in the lungs. Seven patients had DVT that had not been suspected before death. For four patients, PE was the cause of death.¹

• These findings suggest that clinicians should maintain a high index of suspicion for VTE in COVID-19 patients.¹

Patients with severe COVID-19 may have myocardial injury (e.g., elevated troponin, electrocardiogram signs), which may be thrombotic ACS or myocarditis.²

Hemostasis lab abnormalities seen in COVID-19 patients include elevated D-dimer, low platelets, prolonged PT, and shortened aPTT.²

Which COVID-19 inpatients should receive VTE prophylaxis?

All hospitalized COVID-19 patients should receive VTE prophylaxis.^6,19

Use LMWH (or fondaparinux for patients with HIT) for most patients.⁶

• IPC is an alternative if an anticoagulant cannot be used, but combining mechanical and pharmacologic prophylaxis is generally not recommended.^3,6

Should higher-than-usual anticoagulant doses be used for VTE prophylaxis in critically ill COVID-19 patients?

Current data support starting with standard-dose VTE prophylaxis in critically ill COVID-19 patients.¹⁹

• In critically ill COVID-19 patients, a full-dose heparin (mostly enoxaparin) or intermediate-dose LMWH (enoxaparin 1 mg/kg/day) does not improve outcomes (e.g., thrombosis, mortality, need for organ support) vs usual-dose prophylaxis (e.g., enoxaparin 40 mg once daily) [Evidence level B-1].^5,9,12

If, despite prophylactic anticoagulation, COVID-19 patients develop clots in vascular access devices or extracorporeal circuits, consider trying a different anticoagulant, or increasing the dose (i.e., full or intermediate dose) if bleeding risk allows.⁶

Should higher than usual anticoagulant doses be used for VTE prophylaxis in moderately ill hospitalizedCOVID-19 patients?

Current data suggest that VTE prophylaxis using full-dose anticoagulation (preferably LWWH) can benefit moderately ill, select patients (e.g., elevated D-dimer, requiring only low-flow oxygen, non-pregnant), without increased bleeding risk (e.g., platelets <50 x 10⁹/L, hemoglobin <8 g/dL, DAPT, hospital visit for bleeding within the past 30 days, bleeding disorder) [Evidence level B-1].^5,15,16,19

• In a multiplatform adaptive-design trial (MPT) hospitalized patients not needing high-flow oxygen, noninvasive or invasive mechanical ventilation, or vasopressors, full-dose anticoagulation with a heparin (mostly enoxaparin) started within 72 hours of admission or positive in-hospital COVID-19 test increased days free of cardiovascular or respiratory support vs usual-dose VTE prophylaxis.¹⁵ Major bleeding occurred in 1.9% of the full-dose anticoagulation patients vs 0.9% of the usual-dose prophylaxis patients (not a statistically significant difference).
• In the RAPID trial, non-ICU patients with elevated D-dimer (n=465) were randomized to a full-dose or prophylactic-dose heparin (mostly enoxaparin).¹⁶ Although there were fewer VTEs among patients who received full-dose anticoagulation (0.9% vs 2.5%; OR 0.34, 95% CI 0.07 to 1.71, p=0.19), there was no significant difference in occurrence of the primary composite outcome (need for non-invasive or invasive mechanical ventilation, ICU admission, or death). Major bleeding occurred in 1.7% of the full-dose anticoagulation patients vs 0.9% of the usual-dose prophylaxis patients (not a statistically significant difference). Mortality at 28 days was lower in the full-dose arm (OR 0.22 [95% CI 0.07 to 0.65, p=0.006).
• In the HEP-COVID study (n=257), patients with D-dimer >4 times ULN or sepsis-induced coagulopathy score ≥4 were randomized to full-dose enoxaparin or a prophylactic/intermediate-dose heparin (mostly enoxaparin).⁵ The primary composite outcome was VTE, arterial thromboembolism, or death. Almost all patients were enrolled based on their D-dimer level. About 2/3 of the study patients were non-ICU patients. Only non-ICU patients benefited from full-dose enoxaparin, driven by reduced thrombosis (NNT = 5 to prevent one thrombotic event). Major bleeding occurred in 2.4% of the non-ICU patients treated with full-dose enoxaparin, vs 2.3% in the prophylactic/intermediate-dose group.

Patients in MPT and RAPID were relatively young (mean age ~60 years^15,16). Patients with high bleeding risk or dual antiplatelet therapy were excluded from all three studies.^5,15,16 In MPT and RAPID, few included patients were taking a single antiplatelet (~12%), and only about 7% had chronic kidney disease.^15,16 In HEP-COVID, <4% of patients had chronic kidney disease.⁵

Should VTE prophylaxis be considered for outpatients?

It is generally recommended that VTE prophylaxis be discontinued at discharge.^6,19

• Post-discharge, VTE risk is low and comparable to the risk of other medical conditions.⁶ However, COVID-19 patients discharged early to free up hospital beds may need an assessment of their VTE risk.⁶
• In a large, placebo controlled study in non-COVID-19 discharged medical patients (MARINER), extended duration VTE prophylaxis with rivaroxaban did not provide net benefit [Evidence level A-1].¹⁸ NNT to prevent one symptomatic VTE was 430. Major or clinically important bleeding (e.g., requiring discontinuation or medical contact) occurred in 1.7% of rivaroxaban patients vs 0.7% of placebo patients (NNH = 143).
• In COVID-19 patients, one study (MICHELLE) suggests benefit of post-discharge VTE prophylaxis.⁷ However, findings are limited by open-label design, small sample size (n=320), and inclusion of patients with low bleeding risk (e.g., mean age 57.1 years, 95% not taking an antiplatelet, no severe renal disease).⁷ In MICHELLE, rivaroxaban 10 mg/day for 35 days post-discharge prevented a symptomatic or fatal VTE in 1 in 23 COVID-19 patients with high thrombosis risk (IMPROVE score ≥4, or 2-3 with D-dimer >500 ng/mL).⁷ Major bleeding did not occur.⁷

At discharge, educate COVID-19 patients to seek help for symptoms of VTE.⁶

For patients with mild COVID-19 who are isolating at home, advise keeping active.²

How is thromboembolism in COVID-19 patients treated?

Anticoagulation, for at least three months, is the mainstay of treatment.^2,3 Initiate treatment with a parenteral agent in critically ill patients.³

Consider using anti-factor Xa levels to monitor UFH due to aPTT abnormalities in these patients.¹¹

For patients with recurrent VTE despite appropriate anticoagulation, consider switching from oral therapy to LMWH, or increasing the LMWH dose by 25% to 30% in patients failing standard-dose LMWH, based on low-quality evidence in other populations.³

Catheter-directed therapy or systemic thrombolysis should be reserved for the most serious cases.^2,3 See our chart, Pulmonary Embolism: Focus on Thrombolytics, for more information.

Reserve IVC filters for recurrent PE despite appropriate anticoagulation, or clinically important VTE with absolute contraindications to anticoagulation.²

What are some general considerations for antithrombotic use of special relevance to COVID-19?

Extrapolating from other populations, antiplatelets (e.g., aspirin) are likely inferior to anticoagulants for VTE prophylaxis in COVID-19 patients needing hospitalization.³

For patients who might need procedures, consider parenteral antithrombotics over oral antithrombotics due to shorter duration of action.²

DOACs may be difficult to manage in hospitalized COVID-19 patients due to clinical instability resulting in impaired oral drug absorption or deterioration of renal function, and drug interactions with COVID-19 treatments.³

In the hospital, consider fondaparinux (not for intensive care patients) or LMWH over UFH to reduce caregiver viral exposure and to reduce the risk of missed doses; fondaparinux and LMWH require less frequent blood draws for monitoring and fewer daily doses.^2,3 However, UFH might be preferred for patients with hemodynamic instability or renal insufficiency due to quicker offset.³

In patients with ACS and elevated bleeding risk (e.g., due to DIC), consider clopidogrel over other antiplatelets.²

In patients taking antithrombotics chronically who develop known or suspected DIC without overt bleeding, consider risk/benefit of reducing the intensity of therapy or discontinuation. For example, in patients taking DAPT, consider risk/benefit of continuing DAPT if platelets ≥50,000/mm³, switching to a single antiplatelet if platelets are ≥25,000 to <50,000/mm³, or discontinuing if platelets <25,000/mm³.²

In outpatients, consider a DOAC or LMWH over warfarin if home or drive-in INR monitoring is not available, assuming use is feasible given cost, indication (e.g., prosthetic heart valve), comorbidities (e.g., pregnancy), etc.^2,8

Outpatients with COVID-19 and previously diagnosed thrombotic or CV disease should generally continue their usual antithrombotic regimen (e.g., aspirin, anticoagulant).² 

Educate outpatients taking antithrombotics to discern clinically important bleeding from nuisance bleeding to reduce unnecessary emergency room visits.²

Be alert for drug interactions between antithrombotics and drugs used to treat COVID-19. Select drug interactions are covered in the next section.

What are some select drug interactions between anticoagulants and drugs used treat COVID-19?

Dexamethasone (high dose): increased warfarin effect.¹⁴ Dexamethasone is a CYP3A4 inducer, but whether it significantly reduces DOAC efficacy is unknown.⁶

Methylprednisolone (high dose): increased warfarin effect.¹⁴

Paxlovid (nirmatrelvir/ritonavir): increased rivaroxaban effect; avoid.¹⁷ Monitor INR closely in patients taking warfarin (may increase or decrease).¹⁷

Sarilumab: may increase CYP450 activity, potentially decreasing efficacy of warfarin, apixaban, and rivaroxaban.¹⁴

Tocilizumab: may increase CYP450 activity, potentially decreasing efficacy of warfarin, apixaban, and rivaroxaban.¹⁴

See the Liverpool COVID-19 Drug Interaction website: https://www.covid19-druginteractions.org to screen for more drug interactions.